Evaluation of The Composite Action of Cold Formed Steel Buil - 2023 - Thin Walle

Thin-Walled Structures 183 (2023) 110390
Contents lists available at ScienceDirect
Thin-Walled Structures
journal homepage: www.elsevier.com/locate/tws
Evaluation of the composite action of cold-formed steel built-up battened

columns composed of two sigma-shaped sections
Rohola Rahnavard a ,∗, Mahtabsadat Razavi b , Nader Fanaie b , Hélder D. Craveiro a ,∗
a University of Coimbra, ISISE, Department of Civil Engineering, Coimbra, Portugal
b
Department of Civil Engineering, K. N. Toosi University of Technology, Tehran, Iran
ARTICLE INFO ABSTRACT

Keywords: Cold-formed steel (CFS) built-up battened columns are employed to obtain a sufficient load-bearing capacity
Cold-formed steel when individual profiles cannot withstand the specified stresses. This sufficient load-bearing capacity can be
Optima number of fasteners obtained when the individual profiles present a composite action. This paper presents the optimal layout for
Optima distance
CFS built-up battened columns to achieve a high level of composite action between two 𝛴-shaped CFS profiles.
Composite action
A calibrated finite element modeling technique was used to evaluate the effect of the number of fasteners per
Battened columns
Batten panel
batten panel and the distance between the batten panels on the axial capacity of CFS built-up battened columns.
The optimal number of fastener rows and the distance between the batten panels were determined using 100
models, in which a high level of composite action was achieved between CFS profiles. Then the maximum axial
load-bearing capacity of additional 600 finite element models using the optimal layout was compared with the
analytical predictions in the European standard EN1993-1-1. The results recommended four rows of fasteners
per batten panel. The results suggested an optimal distance of 30𝑖𝑚𝑖𝑛 between batten panels. The results showed
a close agreement between finite element and analytical predictions from EN1993-1-1, following the procedure
specified in clause 6.3.1, adopting the modified slenderness ratio from the American Specification AISI S100-16.
Reliability analysis was also performed to evaluate the analytical methods by targeting a reliability index of
2.5. The results showed that the analytical prediction following the EN1993-1-1, clause 6.3.1, incorporated
with the recommended slenderness ratio from the AISI S100-16, is a reliable analytical procedure for CFS
battened columns.
1. Introduction knowledge of how these sections behave is still restricted. Because

CFS built-up sections are composed of individual profiles and batten
As an effective substitute for hot-rolled steel, cold-formed steel plates connected by fasteners, the buckling modes and interactions
(CFS) has been widely used in various structural applications in recent are very different from those seen in individual profiles. Furthermore,
years. Lighter, stronger, more environmentally friendly, and less ex- the distance between the batten panels and the number of fasteners
pensive to maintain are just some benefits CFS-built structures provide per batten panel can impact the prevalent buckling modes and the
over their hot-rolled alternatives. Moreover, the CFS built-up sections actual efficiency of the composite action between the existing profiles.
cover various applications such as columns, beams, and even diagonally Because accounting for partial or full composite action is essential when
strap-braced stud walls [1–4]. The cold-formed steel profiles have working with built-up battened columns, developing tailored design
high strength-to-weight ratios; however, their load-bearing capacity is methods must be used in practice.
relatively low. This low load-bearing capacity is due to their thin walls Despite the various advantages of CFS built-up battened sections,
and sensitivity to local buckling phenomena. CFS profiles’ potential low the available design codes [5,6] rarely provide specialized design in-
load-bearing capacity can be addressed using the possibility of com- structions. Even though EN 1993-1-3 [5] offers a global buckling curve
bining multiple individual shapes known as CFS built-up sections. CFS for calculating the buckling resistance of a limited number of individual
built-up sections are usually created by fastening individual sections and built-up sections, it does not include particular design requirements
together. Another way to create CFS built-up sections is to connect for built-up battened sections. While critical parameters such as the
individual profiles using batten panels. CFS built-up While individual maximum distance between the batten panels and the optimal number
profiles may be used to develop an extensive range of built-up mem- of fasteners per batten panel can significantly influence the behavior
bers with various cross-section shapes and capabilities, fundamental of CFS built-up battened columns, there are no specific criteria to
∗ Corresponding authors.
E-mail addresses: [email protected], [email protected] (R. Rahnavard), [email protected] (H.D. Craveiro).
https://doi.org/10.1016/j.tws.2022.110390
Received 19 July 2022; Received in revised form 12 October 2022; Accepted 20 November 2022
Available online 9 December 2022
0263-8231/© 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
R. Rahnavard, M. Razavi, N. Fanaie et al. Thin-Walled Structures 183 (2023) 110390
Notation 𝛷𝑐 Safety factors for the available axial

strength based on the Limit States Design
A Area of a cross section
(LSD)
𝐴𝑐ℎ Cross-sectional area of one chord
𝜆 Slenderness of a compression element
𝐴𝑒𝑓 𝑓 Effective area of a cross section
𝛼 Imperfection factor
E Modulus of elasticity
𝛾𝑀1 Partial factors for resistance of members
I Second moment of area of a cross section
𝜇 Efficiency factor
𝐼𝑏𝑝 Second moment of area of one batten panel
𝜈 Poisson’s ratio
𝐶𝑝 Correction factor in reliability analysis
𝜒 Reduction factor for relevant buckling mode
𝐼𝑐ℎ Second moment of area of one chord
𝛽 Reliability index
𝐼𝑒𝑓 𝑓 Effective second moment of area of a
𝜙 Resistance factor
cross-section
K Spring stiffness per unit length of a stiffener
L Length of a member
𝐿𝑒 Effective length of a member battened compression members that do not even include the distance
𝑁𝑏,𝑅𝑑 Design buckling resistance of a compression between batten panels and the required fasteners to obtain a full
member composite action.
𝑁𝑐,𝑅𝑑 Design resistance of steel cross-section un- Recently, a substantial study has been conducted examining the
der compression behavior of CFS built-up members [8–27]. Rahnavard et al. [8,9]
𝑁𝑐𝑟 Elastic buckling force numerically investigated the effect of the number of fastener rows
𝑁𝑐𝑟,𝐵 Elastic flexural buckling force of a battened per batten panel and the optimal distance between the batten panels.
compression built-up member Their study was focused on CFS built-up battened columns created
by double C-shaped profiles. They also investigated the effect of the
𝑁𝑐𝑟,𝑒𝑓 𝑓 Elastic flexural buckling force of a battened
geometric shape of the batten panel by comparing a flat batten panel
compression effective built-up member
with a U-shaped one. Their results recommended the optimal layout of
𝑁𝑐𝑟,𝐹 Elastic flexural buckling force
three rows of fasteners per batten panel. A series of built-up specimens
𝑁𝑐𝑟,𝑇 Elastic torsional buckling force
were tested under compression by Dar et al. [10–16] to investigate the
𝑁𝑐𝑟,𝑇 𝐹 Elastic torsional–flexural buckling force spacing effect between individual profiles. The individual lipped chan-
𝑆𝑣 Shear stiffness of a built-up battened mem- nels were arranged back-to-back with specific spacing. They examined
ber the key parameters, such as the maximum load-bearing capacity and
𝑑𝑏𝑝 Distance between batten panels failure modes. Their specimens’ slenderness covered a suitable range
𝑀𝑚 , 𝐹𝑚 , 𝑉𝑀 , Statistical parameters in the reliability anal- in which the local and global failure modes were seen in their tests.
and 𝑉𝐹 ysis They compared their experimental results with the design approach,
including the EN1993-1-3 [5] and the AISI S100 [6], resulting in a more
𝑉𝑄 Coefficient of the variation of the mean load accurate axial load prediction following the EN1993-1-3 [5].
effect in the reliability analysis Reyes and Guzmán [17] assessed the impact of the slenderness ratio
n Number of models in reliability analysis in CFS built-up box sections by testing 48 experimental specimens.
m Degree of freedom in the reliability analysis Their CFS built-up columns were created using two C-shaped profiles
𝑑𝑓 Distance between fasteners in the direction connected by seam welds with varying spacings. Their results indicated
of load transfer that the modified slenderness provided in the AISI S100 [6] does not
necessarily apply to the built-up members with thicknesses of 1.5 and
𝑓𝑦 Yield strength
2 mm and seam weld spacing of less than 600 mm. An experimental
𝑒1 The end distance from the center of the
and numerical study on the axial behavior of CFS built-up section
fastener to the adjacent end of the batten
columns with batten panels was carried out by Dabaon et al. [18–20].
panel, in the direction of load transfer
They compared the experimental and numerical results with analytical
𝑒2 The edge distance from the center of the
predictions from available design codes, including the AISI S100 [6],
fastener to the adjacent end of the batten the Australian/New Zealand Standard [21], and the EN 1993-1-3 [5].
panel, perpendicular to the direction of load Their results showed that the code predictions were unconservative for
transfer those specimens that failed by local mode, while for those that failed
b Width of a cross-section, plate width by global mode, a conservative prediction was achieved.
h Height of a cross-section El Aghoury et al. [22–24] studied the axial behavior of CFS built-
ℎ0 Distance between the centroids of chords up battened columns under concentric and eccentric axial loads. They
𝑛𝑏 Number of batten panels tested battened columns where four CFS angles were fastened using
r Radius of gyration of a cross-section batten panels. They compared their failure load obtained from the tests
𝑟𝑖 Minimum radius of gyration of the full unre- and simulation with analytical prediction according to the EN1993-1-
duced cross-sectional area of an individual 3 [5], resulting in the accuracy of the EN1993-1-3 [5] design procedure.
shape in a built-up member Moreover, their results indicated that the design procedure might have
t Plate thickness errors when the failure mode is due to the interaction of local and
global buckling modes. However, the design predictions are unsafe for
the column with medium slenderness when the interaction between
individual members characterizes the failure modes. A series of ex-
determine them. Moreover, no specific criteria exist to determine the perimental tests and numerical analyses were carried out by Anbarasu
et al. [25], focusing on CFS built-up battened columns made of lipped
effective composite action among CFS profiles connected using batten
angles. They compared the maximum load-bearing capacity obtained
panels. The EN 1993-1-1 [7] offers design suggestions for limited from the tested specimens and numerical models with the analytical
2
prediction following the Australian/New Zealand Standard [21] and as 1.75, 1.50, 1.25, 1.00, and 0.75 (See Fig. 2). These attached batten
the EN1993-1-3 [5]. According to EN1993-1-3 [5], the predictions were panels are spaced differently throughout the column’s length to obtain
safe only for the relatively long CFS built-up battened columns, while the optimal distance between the batten panels. The EN1993-1-1 [7]
unsafe predictions were for the relatively short ones. recommends 15𝑖𝑚𝑖𝑛 and 70𝑖𝑚𝑖𝑛 for limited CFS built-up configurations,
Fratamico et al. [26,27] tested sixteen CFS built-up columns with where 𝑖𝑚𝑖𝑛 is the minimum radius of gyration of an individual CFS
different back-to-back lipped channels. Their results showed that the profile. However, the optimal distance between the batten panels still is
composite action was achieved most likely when more isolated global unclear for the built-up battened column to achieve composite action.
This study investigates the batten panels with a distance of (𝑑𝑏𝑝 ) 15𝑖𝑚𝑖𝑛 ,
buckling occurred. Moreover, their results indicated a higher load-
30𝑖𝑚𝑖𝑛 , 45𝑖𝑚𝑖𝑛 , and 70𝑖𝑚𝑖𝑛 to address the mentioned limitation.
bearing capacity using more fasteners connecting the lipped chan-
nel’s webs. Zhang and Young [28–31] evaluated the CFS built-up
columns using tests and simulations in which the CFS profiles with 2.2. The geometry details
web stiffeners were used. They compared the maximum load-bearing
capacity with the analytical prediction from the AISI S100 [6] and
This paper presents a numerical study on the composite action
the Australian/New Zealand Standard [21], resulting in the excellent
of CFS built-up battened columns. 12 𝛴-shaped CFS profiles were
applicability of the AISI S100 [6]. The applicability of the AISI S100 [6] considered for this study to cover a broader range of available profile
for different types of CFS profiles was investigated in large sets of sizes and, consequently, wider slenderness. The geometry details of
parametric studies [32–36]. A comprehensive study was conducted by individual CFS profiles are shown in Fig. 3. Each built-up battened
Rasmussen et al. [37], which discussed the composite action of built-up column includes two CFS 𝛴-shaped profiles, in which the profiles are
CFS components. They considered the effect of the fasteners’ location connected using the batten panels along the length of the column. The
and effective flexural rigidity to explain the composite action between batten panels are fastened to both 𝛴-shaped profile flanges, as indicated
the individual profiles. Other CFS built-up columns, including built- in Fig. 4. Self-drilling screws were considered with a diameter (d) of
up Double-Z members, were also investigated using experimental and 6.3 mm and a length of 35 mm. The general demand for the end
numerical models [38–56]. distance (𝑒1 ), edge distance (𝑒2 ), and distance between the fasteners
Although some studies were performed recently on CFS built-up (𝑑𝑓 ) were defined according to the requirements provided by EN1993-
battened columns, their behavior still needs to be explored more. 1-3 [5]. According to EN 1993-1-3 [5], the end distance from the center
Additional research related to the design of CFS built-up battened of the fastener to the adjacent end of the batten panel, in the direction
sections is essential to understand their behavior better and develop of load transfer and distance between the fasteners (𝑑𝑓 ) should be
the design procedures. Moreover, further investigations are necessary greater than three times the fastener hole (𝑒1 ≥ 3d and d𝑓 ≥ 3d).
to determine the optimal layout to achieve composite action among CFS Moreover, the edge distance from the center of the fastener to the
profiles. This study evaluates the CFS built-up battened columns (two adjacent end of the batten panel, perpendicular to the direction of load
𝛴-shaped) under axial compression. An extensive set of finite element transfer, should be greater than 1.5 times the fastener hole (𝑒2 ≥ 1.5d).
models were simulated using the Abaqus [57]. Seventeen experimental As shown in Fig. 4, the distance between the fasteners (𝑑𝑓 ) was 20 mm
specimens of CFS built-up battened columns were selected from the for all models. The end distance from the center of the fastener to the
adjacent end of the batten panel was considered 25 mm and 21.5 mm
literature to validate the numerical models. The optimal number of
in the direction of load transfer (𝑒1 ) and perpendicular to the direction
fasteners per batten panel and the distance between the batten panels
of load transfer (𝑒2 ).
were determined using static displacement control analysis on 100
Different column lengths, including 2000 mm, 3000 mm, 4000 mm,
models. Then, 600 models with different lengths and spacings using
5000 mm, and 6000 mm, were considered for each CFS built-up bat-
the optimal configuration are presented. The load-bearing capacities of
tened column. Two boundary conditions, including pinned and fixed,
these 600 finite element models are compared with the design predic-
were investigated in this study. Table 1 lists the information on the
tions according to EN 1993-1-3 [5]. The available design procedures in
models utilized in this study, including their identifying codes.
EN 1993-1-3 [5] are discussed and modified.
2.3. Numerical modeling

2. Description of the research study
2.3.1. Material properties
2.1. The concept of the present study S280GD+Z steel was utilized in this work for both CFS 𝛴-shaped
profiles and batten panels. This study considered the mechanical char-
acteristics of S280GD+Z steel conducted by Craviero et al. [58–61]. The
This paper aims to present the optimal number of fastener rows and stress–strain curve obtained from tensile coupon testing on S280GD+Z
the distances between the batten panel, representing a composite action steel [58–61] is shown in Fig. 5. The CFS material was defined using
between the individual CFS 𝛴-shaped profiles. The built-up battened the Abaqus elastoplastic model [57]. Young’s modulus and Poisson’s
columns were assembled using two CFS 𝛴-shaped profiles connected ratio are required to model steel materials’ elastic characteristics. Ac-
using batten panels. The batten panels are fastened to the flanges of cording to the tensile coupon test [58–61], Young’s modulus was 204
the CFS 𝛴-shaped profiles. The optimal number of the fastener rows GPa. Poisson’s ratio was assumed to be 0.3. The plastic behavior was
per batten panel to reach a composite action between two CFS profiles defined using a true stress–strain curve. The true stress–strain curve
is unclear. Therefore, the effect of the number of fasteners rows was was converted using the plastic range of the engineering stress–strain
considered by examining 1 row to 6 rows of fasteners per batten, as curve (Fig. 5). Note that isotropic hardening was chosen to determine
shown in Fig. 1. These CFS 𝛴-shaped profiles can be positioned at the plastic’s characteristics. According to the tensile coupon test [58–
various spacings; as a result, different geometries and load-bearing 61], the proportional stress limit was 212.5 MPa. Moreover, the yield
capacities depending on the design requirement, can be achieved. This stress was 306.81 MPa. The ultimate stress and associated strains were
study covers different profile spacings by considering five ℎ∕𝑏 ratios 424.04 MPa and 20%, respectively.
3
Fig. 1. Example of a battened column with (ℎ∕𝑏 = 0.75) and various fasteners rows per batten panel. (a) one row, (b) two rows, (c) three rows, (d) four rows, (e) five rows, and
(f) six rows.
Fig. 2. The battened columns with varied ℎ∕𝑏 ratio.
4
Fig. 3. Geometry details of individual CFS profiles.
Fig. 4. Geometric details of the battened column.
5
Table 1
The finite element models.
Models series Number of Column length, Batten panel Batten panel ℎ∕𝑏 Number of
models L (mm) distance, 𝑑𝑏𝑝 (i𝑚𝑖𝑛 ) thickness (mm) fasteners rows
2𝛴150 × 43 × 2 50 2000 to 6000 30 2 1.75, 1.50, 1.25, 1.00, and 0.75 4
2𝛴150 × 43 × 2.5 50 2000 to 6000 30 2 1.75, 1.50, 1.25, 1.00, and 0.75 4
2𝛴150 × 43 × 3 50 2000 to 6000 30 2 1.75, 1.50, 1.25, 1.00, and 0.75 4
2𝛴170 × 43 × 1.5 50 2000 to 6000 30 2 1.75, 1.50, 1.25, 1.00, and 0.75 4
2𝛴170 × 43 × 2 50 2000 to 6000 30 2 1.75, 1.50, 1.25, 1.00, and 0.75 4
2𝛴170 × 43 × 3 50 2000 to 6000 30 2 1.75, 1.50, 1.25, 1.00, and 0.75 4
2𝛴200 × 43 × 1.5 50 2000 to 6000 30 2 1.75, 1.50, 1.25, 1.00, and 0.75 4
2𝛴200 × 43 × 2 150 2000 to 6000 15, 30, 45, 70 2 1.75, 1.50, 1.25, 1.00, and 0.75 1, 2, 3, 4, 5, and 6
2𝛴200 × 43 × 3 50 2000 to 6000 30 2 1.75, 1.50, 1.25, 1.00, and 0.75 4
𝛴220 × 43 × 1.5 50 2000 to 6000 30 2 1.75, 1.50, 1.25, 1.00, and 0.75 4
𝛴220 × 43 × 2 50 2000 to 6000 30 2 1.75, 1.50, 1.25, 1.00, and 0.75 4
𝛴220 × 43 × 3 50 2000 to 6000 30 2 1.75, 1.50, 1.25, 1.00, and 0.75 4
was defined according to the stress–strain curve presented for

self-drilling [62,63].
- A combination of Cartesian connector and fastener: the Cartesian
connector specifies the connection between two nodes on two
surfaces in this approach. In this modeling approach, the shear
stiffness of the self-drilled fastener was defined according to the
experimental tests conducted by Huynh et al. [62].
2.3.3. Imperfection, boundary conditions, and loading

Fig. 6 shows the details of boundary conditions and loading. Two
reference points, RP-1 and RP-2, were coupled to the column ends.
Two boundary conditions, including pinned and fixed, were considered
in this study. As is shown in Fig. 6, only three transitional directions
(𝑈x , 𝑈y , and 𝑈z ) were restrained from defining the pinned boundary
condition. The transitional direction (𝑈x , 𝑈y , and 𝑈z ) and the rotational
movements (𝜃x , 𝜃y , and 𝜃z ) were also blocked to define the fixed
boundary condition. However, it should be mentioned that one of the
reference points (RP-2) was free to move transitionally in the longitu-
dinal direction to be axially loaded. No relative movement between the
Fig. 5. Stress–strain diagram of the S280GD+Z [58–60].
individual elements of the built-up member could occur because both
column ends were coupled to the reference points.
Linear buckling analysis and nonlinear general static analysis avail-
able in the Abaqus [57] were used in this study. The linear buckling
analysis was undertaken to obtain the buckling mode shapes. These
2.3.2. Interactions and meshing buckling mode shapes were used to define the initial imperfection for
Proper interactions to specify the contact between batten panels and
the CFS profiles. A combination of local and flexural buckling modes
profiles are critical parameters for successful numerical modeling. The
was proposed to determine the initial imperfection. Previous research
surface-to-surface interaction was selected for defining the interaction
studies suggested the imperfection amplitude of 1/1000 of the column
among all parts. Contact properties were defined, including normal
(hard contact) and tangential (coefficient of 0.2). length for global buckling mode. Craveiro et al. [60] proposed a value
Three modeling techniques were employed to model the self-drilled of 1/200 of the section height as amplitude for the local buckling mode.
fasteners: However, the readers should be aware that EN 1993-1-1 [7], clause
6.4.3, recommended L/500 for the battened columns.
- A combination of beam connectors and fasteners: the beam con-
Moreover, a sensitivity analysis was performed by Rahnavard et al.
nector specifies the connection between two nodes on two sur-
[8,9] that showed an accuracy of axial load prediction by selecting
faces in this approach. From one surface to another, the ‘‘fastener’’
tool attached the actual radius of the fasteners to the beam 1/1000 of the column length for the global buckling mode and 1/200
connection. Fig. 6 illustrates a combined ‘‘beam connector and of the section height as amplitude for the local buckling mode. Fig. 7
fastener’’. shows examples of selected buckling modes for the 2𝛴200 × 43 × 2
- Modeling the actual self-drilled fasteners using solid elements: for model. The defined initial geometric imperfection modes depicted in
this modeling technique, the elastic and plastic material property Fig. 7 include global buckling (Fig. 7-a) and local buckling (Fig. 7-b).
6
Fig. 6. Finite element modeling.
Fig. 7. The used buckling modes as initial imperfection; (a) global and (b) Local.
7
batten panels along the columns was 175 mm. Moreover, The spacing
between the U-shaped CFS was 0 mm and 100 mm for T2T-0-175 and
T2T-100-175, respectively.
Fig. 8 compares the axial load-bearing capacity obtained from the
experimental tests [10] and numerical models with different mod-
eling approaches. For the case of T2T-0-175, the axial load-bearing
capacity obtained from the experimental test was 71.5 kN. The nu-
merical results showed 75.5 kN, 74.2 kN, and 68.58 kN for ‘‘beam
connector+fastener’’, solid element, and ‘‘beam connector+fastener’’,
respectively. By comparing the results for T2T-0-175, it can be seen
that all modeling approaches predicted the experimental results well.
A similar conclusion was obtained for the case of T2T-0-175. The
axial load-bearing capacity obtained from the experimental test was
163.01 kN. The numerical results showed 164.13 kN, 163.98 kN, and
161.94 kN for ‘‘beam connector+fastener’’, solid element, and ‘‘beam
connector+fastener’’, respectively. Therefore, it can be concluded that
the simplified connector+fastener technique can represent the self-
drilling fastener behavior as accurately as the actual solid element. In
addition, modeling the actual fastener using solid elements required
much computational calculation and was a time-consuming modeling
Fig. 8. Axial force for experimental tests [10] and numerical modeling with different
approach. Consequently, the simplified connector+fastener technique
modeling approaches.
was used in the following.
Additional experimental tests were selected and modeled using the
mentioned ‘‘beam connector+fastener’’ techniques. Five CFS battened
column specimens with various lengths from a research study con-
ducted by Dar et al. [11] were modeled. The selected CFS built-up
battened column specimens were with a lengths of 600 mm (BS-2-600),
1000 mm (BS-2-1000), 1500 mm (BS-2-1500), 2000 mm (BS-2-2000),
and 2400 mm (BS-2-2400). The axial force vs. vertical displacement
curves obtained from the finite element models and experimental spec-
imens were compared (Fig. 9). As shown in Fig. 9, the finite element
models predicted the axial capacity of the experimental CFS built-up
battened columns.
Two more built-up CFS experimental column specimens [61] with
rectangular (R-2𝛴+2U) and square (S-2𝛴+2U) section shapes were
considered to verify the numerical models employed in this work. The
experimental CFS built-up column specimens were built using CFS 𝛴-
shaped and U-shaped with a thickness of 1.5 mm. The length of the
selected CFS built-up column specimens was 1050 mm. More infor-
mation about the experimental specimens is available in [61]. Fig. 10
compares the experimental and numerical results for the rectangular
CFS built-up column in load-bearing capacity and buckling deforma-
tions. Very similar load-bearing capacity and deformation can be seen,
indicating excellent agreement between experimental and numerical
results. A similar conclusion can be made for the square CFS built-up
column, as shown in Fig. 11.
Ten more built-up CFS experimental column specimens composed
of two 𝛴-shaped CFS profiles [30] were considered to investigate
the reliability of the modeling techniques used in this paper. The
Fig. 9. Force vs. shortening displacement curves for experimental specimens [11] and experimental CFS built-up box column specimens were built using two
numerical models.
CFS with edge and web stiffeners. Two thicknesses, including 0.48 mm
and 1.00 mm, were considered according to the experimental study
by Zhang et al. [30]. Different lengths were selected for CFS built-
2.3.4. Finite element modeling validation
up box columns, including 300 mm, 800 mm, 1400 mm, 2000 mm,
In this section, previous studies’ experimental tests were selected
and 3200 mm. More information about the experimental specimens is
and modeled using the mentioned techniques in Section 2. Two CFS
available in [30]. A comparison of the maximum axial load obtained
built-up battened columns (T2T-0-175 and T2T-100-175) were selected
from the experimental work conducted by Dar et al. [10] to compare from experimental specimens and numerical models is listed in Table 2.
the modeling techniques used for self-drilled fasteners. Two U-shaped The experimental tests [24] and the numerical models showed a similar
CFS profiles were connected in the experimental specimens by employ- load-bearing capacity, indicating excellent agreement between exper-
ing batten panels of 6 mm thickness and 5 mm diameter self-drilling imental and numerical results (≤7%). These close agreements show
screws. The U-shaped CFS profiles had a web depth of 100 mm, a flange that presented finite element modeling techniques can be reliable for
width of 25 mm, and a thickness of 2 mm. The distance between the performing this parametric study.
8
Fig. 10. Comparison of the experimental [61] and numerical results for rectangular built-up 𝛴-shaped.
Fig. 11. Comparison of the experimental [61] and numerical results for square built-up 𝛴-shaped.
Table 2
Comparison of the test [30] and modeling axial capacities.
Specimen ID Axial load-bearing capacity (kN) FEM∕EXP Specimen ID Axial load-bearing capacity (kN) FEM∕EXP
EXP FEM
VT0.48L300 43.30 45.2 1.04 VT1.0L300 137.6 138.19 1.00
VT0.48L800 41.20 43.05 1.04 VT1.0L800 135.2 131.47 0.97
VT0.48L1400 39.50 41.75 1.06 VT1.0L1400 129.8 123.82 0.95
VT0.48L2000 37.70 39.15 1.03 VT1.0L2000 127.9 119.95 0.93
VT0.48L3200 31.70 34.05 1.07 VT1.0L3200 81.9 79.15 0.97
Mean value 1.04 Mean value 0.96
Standard deviation 0.01 Standard deviation 0.02
9
3. Results and discussion four, the load-bearing capacity increased from 255.28 kN to 294.97
kN, indicating that the axial capacity improved by 15%. However,
3.1. The optimal number of fasteners per batten panels the increase in axial load-bearing capacity is negligible (less than 4%)
when the fasteners row per batten panel rises from four to five and
The effect number of fastener rows is investigated in this section. six. A similar comparison for the pinned battened column is visible in
For this purpose, the force versus displacement curve for CFS built- Fig. 12b. When the h/b ratio was 1.75, as shown in Fig. 13, the axial
up battened columns with a different number of fastener rows varying load-bearing capacity of the battened column with fixed ends (Fig. 13a)
from one row to six rows were compared in Figs. 12 and 13. The and pinned ends (Fig. 13b) increased with the number of rows of
CFS built-up battened column of 2𝛴200 × 43 × 2 with a length fasteners. Fig. 13a demonstrated a 14% improvement in axial capacity
of 3000 mm was chosen for this comparison. The distance between when the number of fastener rows in a battened column with fixed ends
batten panels was considered L/4 for models. Although the axial load- was increased from one to four. However, going from four to five or
bearing capacity analysis was performed for all h/b ratios (Fig. 2), six fasteners per batten panel only results in a little boost in axial load-
this section presents the axial force–displacement curve for the models bearing capacity. Comparing Figs. 12 and 13 shows that the differences
with ℎ∕𝑏 = 0.75 and 1.75, and the maximum capacity of the other between the capacity of the fixed and pinned battened columns are
models was compared in Fig. 14. As seen for the battened column with remarkable when the h/b ratio was 1.75, while for ℎ∕𝑏 = 0.75, this
both fixed ends (Fig. 12a) and pinned end (Fig. 12b) for ℎ∕𝑏 = 0.75, difference was ignorable, resulting in less composite action between
the axial load-bearing capacity increased by increasing the number of CFS profiles when their spacing increased.
fastener rows. As shown in Fig. 12a, when the number of fasteners A summary of the maximum axial load of the battened column
rows of the battened column with fixed ends increased from one to with different h/b ratios is shown in Fig. 14 to investigate the effect
Fig. 12. Force–shortening displacement curves for a battened column with ℎ∕𝑏 = 0.75 and the different number of fastener rows per plate; (a) fixed ends and (b) pinned ends.
Fig. 13. Force–shortening displacement curves for a battened column with ℎ∕𝑏 = 1.75 and the different number of fastener rows per plate; (a) fixed ends and (b) pinned ends.
10
Fig. 14. The effect of the number of fastener rows per batten panel; (a) fixed ends and (b) pinned ends.
Fig. 15. The effect of the number of fastener rows with (a) the same batten panel dimension and (b) different batten panel dimensions.
of the number of fastener rows for more cases of the battened column tests, there were no significant shear deformations along the columns’
geometry. The h/b ratios are explained in Fig. 2 and Table 1. Both length or boundary conditions [30,61]. In addition, no significant
fixed and pinned boundary conditions were evaluated. This comparison shear deformation was reported in the experimental built-up columns
is for the CFS built-up battened column of 2𝛴200 × 43 × 2 with a with battened panels [10,11]. Moreover, this study’s simulation of the
length of 3000 mm and batten panel distance of L/4 (725 mm). As boundary condition involved coupling each column end to a reference
can be seen, by increasing the number of fastener rows up to four per point (RF1 and RF2). Since this is the case, the nodes at the column ends
batten panel, the load-bearing capacity increased remarkably; however, must move simultaneously. At the boundary conditions of the profiles,
with more than four rows of fasteners, the axial load improvement there are no relative deformations (shear deformations). For the same
is ignorable. Therefore, the four rows of fasteners were selected as reason, batten panels close to the column end (50 mm) were used to
the optimal number of fasteners. This study suggests that the more avoid shear deformations.
fastener rows there are, the closer the composite action’s behavior is It should be noted that the width of the battens was increased to
to be detected. It is worth mentioning that the previous study [4] for facilitate the inclusion of a larger number of fasteners. Therefore, a
CFS built-up sections with different profile shapes (C-shaped profiles) more detailed study was performed to discuss further the effect of
suggested the battened column with three rows of fasteners per batten increasing the batten’s stiffness and the number of fasteners. For this
panel as the optimal layout. purpose, the CFS built-up battened column of 2𝛴200 × 43 × 2 with
It should be noted that the shear forces of the fasteners were a length of 3000 mm was chosen for this comparison. The distance
always less than the shear strength of the self-drilled fasteners (12 between batten panels was considered L/4 for models. Fig. 15a shows
kN, according to [62]). Furthermore, for the chosen experimental the axial load ratio for battened columns when the number of rows of
11
fasteners increases from 2 to 6, and the plate width (𝑎𝑏𝑝 in Fig. 4) was to make a composite action between two 𝛴200 × 43 × 2 profiles.
considered 150 mm. Similarly, Fig. 15b shows the axial load ratio for The batten panels’ distances vary by 15𝑖𝑚𝑖𝑛 , 30𝑖𝑚𝑖𝑛 , 45𝑖𝑚𝑖𝑛 , and 70𝑖𝑚𝑖𝑛 .
battened columns when the number of fasteners rows increases from 2 For the case of the battened columns with h/b of 0.75 and fixed ends
to 6, and the plate width increases from 75 mm to 150 mm to facilitate (Fig. 16a), when the batten panel distance is 70𝑖𝑚𝑖𝑛, the capacity is
the inclusion of a more significant number of fasteners. 254.34 kN. By decreasing the distance to 30𝑖𝑚𝑖𝑛 , the axial capacity
As shown in Fig. 15a, by increasing the number of fastener rows increases to 339.17 kN, showing a 33.3% enhancement. However, by
and keeping the batten panel width (𝑎𝑏𝑝 ) the same, the axial capacity of reducing the distance by 50% (15𝑖𝑚𝑖𝑛 ), the axial capacity increases only
the battened columns was increased. However, the axial capacity of the 1.5%. Similarly, for the battened column with h/b of 0.75 and pinned
battened columns was increased more significantly when the number ends (Fig. 16b), the decreasing distance between batten panels up to
of fastener rows and the width of the batten panels were increased 30𝑖𝑚𝑖𝑛 increases the axial bearing capacity significantly. More reduction
(Fig. 15b). Therefore, it is concluded that increasing the axial load- of batten panel distance than 30𝑖𝑚𝑖𝑛 slightly affects the axial behavior.
bearing capacity due to the number of fasteners also affected the batten For the case of the battened columns with h/b of 1.75 and fixed
panel’s stiffness. ends (Fig. 17a), when the batten panel distance is 70𝑖𝑚𝑖𝑛, the capacity
is 254.09 kN. By decreasing the distance to 30𝑖𝑚𝑖𝑛 , the axial capacity
3.2. The optimal batten panels distance increases to 325.01 kN, showing a 27.9% enhancement. However, by
reducing the distance by 50% (15𝑖𝑚𝑖𝑛 ), the axial capacity increases only
The axial force vs. displacement curves for battened CFS built-up 0.5%. Similarly, for the battened column with h/b of 0.75 and pinned
columns with various batten panel distances are depicted in Figs. 16 ends (Fig. 17b), the decreasing distance between batten panels up to
and 17. Note that the column length is 3000 mm, and the h/b is 0.75 30𝑖𝑚𝑖𝑛 increases the axial bearing capacity significantly. More reduction
and 1.75. Four rows of fasteners per batten panel were used, aiming of batten panel distance than 30𝑖𝑚𝑖𝑛 slightly affects the axial behavior.
Fig. 16. Axial force vs. displacement curves for the battened column with ℎ∕𝑏 = 1.75 and various batten panel distances (a) fixed column ends and (b) pinned column ends.
Fig. 17. Axial force vs. displacement curves for the battened column with ℎ∕𝑏 = 1.75 and various batten panel distances (a) fixed column ends and (b) pinned column ends.
12
Fig. 18. The effect of the batten panel distance of battened column with: (a) fixed ends and (b) pinned ends.
Comparing Figs. 16 and 17 shows that the differences between the

capacity of the fixed and pinned battened columns are remarkable
when the h/b ratio was 1.75, while for ℎ∕𝑏 = 0.75, this difference was
ignorable, resulting in less composite action between CFS profiles when
their spacing increased.
For more evaluation of the effect of batten panel distance on the
compression capacity of battened columns, various h/b ratios following
Fig. 2 and Table 1 were analyzed. A comparison of the maximum
capacity of battened columns with different h/b ratios and various
batten panel distances is shown in Fig. 18. For battened columns with
fixed ends (Fig. 18a), the axial load-bearing capacity increases by
decreasing batten panel distance, showing a significant effect.
Moreover, the batten panel distance considerably impacts the pined
battened columns’ axial compression capacity (Fig. 18b). Therefore,
it is concluded that the optimal batten panel distance for the investi-
gated configuration is 30𝑖𝑚𝑖𝑛 . This finding agrees with the suggestion
provided by Rahnavard et al. [8,9] for the battened CFS columns
containing two C-shaped profiles.
The EN1993-1-1 [7] specified a maximum distance of 70𝑖𝑚𝑖𝑛 and
15𝑖𝑚𝑖𝑛 between connectors for various built-up designs. This paper
Fig. 19. The effect of the stiffness value on the axial load-bearing capacity.
proposed a minimum batten panel distance of 30𝑖𝑚𝑖𝑛 and four rows
of fasteners per batten panel for CFS built-up column sections to stiffness. For the fixed column, by increasing the fasteners’ stiffness
accomplish composite action. Note that this suggestion is only valid from 5.5 kN/mm to 25 kN/mm, the axial capacity increased from
for the sections under this investigation. More investigation, including 233.51 kN to 319.47 kN, indicating a 36.8% improvement. However,
experimental tests and numerical modeling of different geometries and by increasing the fasteners’ stiffness from 25 kN/mm to 55 kN/mm,
slenderness, is essential to obtain a reliable conclusion for a wider range the axial capacity increased only 0.4%. A similar conclusion can be
of cross-sections. achieved for the pinned CFS battened columns.
3.3. The effect of the stiffness of the self-drilled fastener 3.4. Failure mode
This section discussed the effect of self-drilling stiffness. For this This section discussed the obtained failure mode for the CFS built-up
purpose, four different stiffness values ranging from 5.5 kN/mm to battened columns. For this purpose, the failure model for the battened
55 kN/mm were considered according to the experimental study con- column of 2𝛴200 × 43 × 2 with the distance between batten panels of
ducted by Huynh et al. [62]. The CFS built-up battened column (two 30𝑖𝑚𝑖𝑛 was selected as suggested in Section 3.2. The column length of
𝛴200 × 43 × 2 profiles) with a length of 3000 mm and an h/b ratio 2000 mm, 3000 mm, 4000 mm, 5000 mm, and 6000 mm was selected.
of 0.75 was considered. The h/b ratio of 0.75 was selected because the Each batten panel fastened the CFS profiles using four rows of fasteners
shear force effect for the battened column with larger spacing between as recommended in Section 3.1. The h/b ratio of 1.75 with pinned
CFS profiles (h/b ratio of 0.75) is more considerable than those with boundary conditions has been considered in this section. Fig. 20 shows
h/b ratios of 1.75, 1.50, 1.25, and 1.00. Fig. 19 compares the effect the failure mode for each CFS built-up battened column with different
of the stiffness of the self-drilled fasteners for fixed and pinned CFS lengths.
battened columns. As can be seen, by increasing the stiffness of the self- Flexural and local buckling about the minor axis was the major
drilling fasteners, the axial load-bearing capacity increased—however, buckling mode for the CFS battened column with lengths of 2000 mm
this increase is remarkable up to a certain level of the self-drilling (see Fig. 20a). The buckling deformation at the maximum load results
13
Fig. 20. Failure mode for the CFS built-up battened column with an h/b ratio of 1.75 and a length of (a) 2000 mm, (b) 3000 mm, (c) 4000 mm, (d) 5000 mm, and (e) 6000 mm.
from the combination of flexural and local deformations. One of the 4.2. Reliability analysis
𝛴-shaped profiles had its web moved inward, causing a distortional
buckling in the flanges. Reliability analysis was carried out in this research to compare pre-
All other models of CFS battened columns with lengths more than dictions obtained from different analytical methodologies. Reliability
2000 mm failed due to global flexural buckling about the minor axis analysis represents the probability of failure. This study evaluated the
(Fig. 20b–e). Flexural buckling about the minor axis, positioned be- analytical methods by comparing their reliability index (𝛽) following
tween the batten panels at mid-span, is the most common failure mode, the AISI S-100 [6]. The reliability index (𝛽) can be obtained from Eq1.
as depicted in Fig. 20b–e. For the case of the CFS column with a length The statistical parameters in Eq. (1), including 𝑀𝑚 , 𝐹𝑚 , 𝑉𝑀 , and 𝑉𝐹
of 3000 mm, a flexural with a minor local buckling was seen (Fig. 20b). were assumed as 1.10, 1.00, 0.10, 0.05, respectively [6,58,61]. 𝑃𝑚 and
There was no evidence of local buckling at the maximum load for the 𝑉𝑃 are the mean value and the coefficient of variation obtained from
columns with lengths of 4000 mm, 5000 mm, and 6000 mm (Fig. 20c– the finite element to the design prediction ratios (see Tables 3 to 5).
e). However, some local deformations are seen at significant lateral The coefficient of the mean load effect variation (𝑉𝑄 ) was considered
deformations (at the end of loading). 0.21 by adopting the dead load (DL) over live load (LL) as 0.2. The
load combination of 1.35DL+1.5LL from Eurocode was selected. The
resistance factor (𝜙) of 1.00 [5] also was considered to calculate the
4. Parametric study and analytical approaches 𝑅
ratio 𝑄 𝑛 . The correction factor (𝐶𝑃 ) was considered for the number
𝑚
of models evaluated in this study. The correction factor (𝐶𝑃 ) can be
4.1. Parametric study calculated using Eq. (2), where n is the number of the models and m
is the degrees of freedom and equal n − 1. The reliability index of 2.5
The optimal distance between batten panels and the number of was targeted for a reliable analytical prediction.
fastener rows per batten panel for a battened column configuration 𝑅𝑛
with double 𝛴-shaped profiles to have a full or partial composite action ln(𝑀𝑚 𝐹𝑚 𝑃𝑚 × 𝑄𝑚
)
𝛽= √ (1)
was discussed in Section 3. Section 3 suggested the optimal distance 2 +𝑉2+𝐶 𝑉2+𝑉2
𝑉𝑀 𝐹 𝑃 𝑃 𝑄
between batten panels as 30𝑖𝑚𝑖𝑛 . Moreover, four rows of fasteners per
batten panel were recommended as the optimal number of fasteners 𝐶𝑃 = (1 + 1∕𝑛)(𝑚∕𝑚−2) (2)
rows. This section uses 600 finite element models and compares them
with analytical approaches. Twelve battened column sections were 4.3. EN1993-1-1, clause 6.3.1 [1]
selected. All models were in lengths, including 2000 mm, 3000 mm,
4000 mm, 5000 mm, and 6000 mm. The h/b ratios, including 1.75, The axial load-bearing capacity of the CFS built-up battened
1.50, 1.25, 1.00, and 0.75, were also considered to expand the results columns can be carried out through advanced design calculations such
for a wider range of spacing. The details of the models can be found in as the finite element and standardized design methods. This section
Table 1. compares the axial load-bearing capacity obtained from finite element
14
Table 3
Reliability analysis for the analytical prediction based on the EN1993-1-1, clause 6.3.1 [7] formulas.
Cold-formed battened columns Fixed battened column Pinned battened column
𝑁𝐹 𝐸𝑀 𝑁𝐹 𝐸𝑀 𝑁𝐹 𝐸𝑀 𝑁𝐹 𝐸𝑀
𝑁𝑏,𝑅𝑑
(curve a) 𝑁𝑏,𝑅𝑑
(curve b) 𝑁𝑏,𝑅𝑑
(curve b)
Number of models # 300 300 300 300

Mean (𝑃𝑚 ) 1.08 1.12 1.05 1.13
Standard deviation 0.051 0.054 0.065 0.065
COV (𝑉𝑃 ) 0.047 0.048 0.062 0.058
Resistance factor (𝜙) 1.00 1.00 1.00 1.00
Correction factor (𝐶𝑃 ) 1.01 1.01 1.01 1.01
Reliability index (𝛽) 2.28 2.42 2.13 2.44
Table 4
Reliability analysis for the analytical prediction based on the EN1993-1-1, clause 6.3.1 [7] formulas incorporated with the recommended
slenderness ratio from the AISI S100-16 [6].
𝑁𝑏,𝑅𝑑
(curve b)

Mean (𝑃𝑚 ) 1.16 1.23 1.13 1.23
COV (𝑉𝑃 ) 0.045 0.047 0.057 0.054
Table 5
Reliability analysis for the analytical prediction based on the EN1993-1-1, clause 6.4.3 [7] formulas.
𝑁𝑏,𝑅𝑑
(curve b)

Mean (𝑃𝑚 ) 1.10 1.13 1.08 1.15
COV (𝑉𝑃 ) 0.054 0.068 0.011 0.010
models and analytical prediction following the EN1993-1-1, Clause load (according to EN1993-1-1, clause 6.3.1 [7]) for various slender-
6.3.1 [7]. ness is shown in Fig. 22. As shown in Fig. 22a, the design buckling
According to the EN1993-1-1, Clause 6.3.1 [7], the design resistance load considering buckling curve a is slightly unconservative, while a
of the class-4 section under pure compression can be calculated by safe prediction is achieved by considering buckling curve b. Although
Eq. (3). Eq. (3) is directly related to the effective sectional area, cal- a slightly unconservative prediction was seen for a few models, the
culated according to [5,64]. The design resistance of a class-4 member overall mean value for the design prediction (1.08) shows an 8%
under pure compression can be calculated by multiplying a reduction difference on the safe side (Table 3). For the case of pinned battened
factor by the design resistance of the section as Eq. (4). The reduction columns (Fig. 22b), the design buckling load prediction considering
factor (𝜒) is calculated by Eq. (5), which is related to the design buckling curve a is unsafe. At the same time, a safe prediction with
buckling curve (𝛼) in Eq. (6) and the slenderness of the compression an average of 13% is achieved by considering buckling curve b.
element (𝜆). The slenderness is obtained using Eq. (7), which depends Table 3 lists detailed data for the reliability analysis for the ana-
on the effective section area and the minimum value among the elastic lytical prediction according to EN1993-1-1, clause 6.3.1 [7]. As can
flexural, torsional, and torsional–flexural buckling forces. The elastic be seen, the correction factor is 1.01, showing a suitable number of
flexural buckling force is calculated using Eq. (8), where 𝐿𝑒 is the total models to perform reliability analysis. The reliability index value for
column length for the pined column and 50% of the column length for the analytical approach according to the EN1993-1-1, clause 6.3.1 [7]
the fixed column. shows values less than 2.5 replicating an unreliable methodology.
A comparison was made between the obtained results from the However, it is seen that the design prediction considering the buckling
finite element modeling and analytical formulas (EN1993-1-1, clause curve b represents a higher value of the reliability index, showing less
6.3.1 [7]) for the fixed (Fig. 21a) and pinned (Fig. 21b) CFS built-up unreliability than those considering the buckling curve a.
battened columns. Fig. 21 showed that the obtained results covered Aef f fy
buckling curve a and above curve b, leading to a safe design by consider- Nc,Rd = (3)
𝛾M1
ing buckling curve b. However, it should be mentioned that the results
for columns with intermediate slenderness (approximately 0.75 ≤ 𝜆 ≤ Nb,Rd = 𝜒Nc,Rd (4)
( √ )
1.5), where the failure occurs due to the interaction between global 2
𝜒 = 1∕ 𝛷 + 𝛷2 − 𝜆 (5)
and local is under the curve a, where is in agreement with previous
studies in [18,19]. Moreover, it can be seen that the battened columns [ ( ) 2
]
using the recommended batten panel distance (30𝑖𝑚𝑖𝑛 ) and four rows 𝛷 = 0.5 1 + 𝛼 𝜆 − 0.2 + 𝜆 (6)
of fasteners represent a full or close to full composite action and can √
be designed with the general methodology in the EN1993-1-1, clause 𝐴𝑒𝑓 𝑓 𝑓𝑦 { }
𝜆= ; 𝑁𝑐𝑟 = min 𝑁𝑐𝑟,𝐹 , 𝑁𝑐𝑟,𝑇 , 𝑁𝑐𝑟,𝐹 𝑇 (7)
6.3.1 [7]. The ratio of the finite element model to the design buckling 𝑁𝑐𝑟
15
Fig. 21. Comparison between the obtained results from the finite element and analytical methods (EN1993-1-1, clause 6.3.1 [7]) for the battened columns with boundary conditions:
(a) fixed and (b) pinned.
Fig. 22. Comparing axial buckling resistance obtained from the finite element models and the EN1993-1-1, clause 6.3.1 [7] for the battened columns with boundary conditions:
member.
𝜋 2 𝐸𝐼 𝜋 2 𝐸𝐴 √
𝑁𝑐𝑟,𝐹 = = ( )2 (8) ( ) ( )2 ( )2
𝐿𝑒2 𝐾𝐿 𝐾𝐿 𝑑𝑏𝑝
𝐿𝑒 ∕𝑟 = + (9)
𝑟 𝑚 𝑟 0 𝑟𝑖
Considering the recommended slenderness ratio, the EN1993-1-1,
4.4. The EN1993-1-1, clause 6.3.1 [7] incorporated the recommended clause 6.3.1 [7] formulations (Eqs. (3)–(8)) can be replaced by Eqs.
slenderness ratio from the AISI S100-16 [6] (10)–(14). The applicability of the EN1993-1-1, clause 6.3.1, by adopt-
ing the slenderness from the AISI S100 [6] was proposed in [8]. This
The North American Specification (NAS) AISI S100 [6] recommends combination successfully predicted the axial load-bearing capacity of a
increasing the slenderness ratio of the battened columns according to battened column made of C-shaped profiles.
Eq. (9). In the recommended slenderness by the AISI S100 [6], the
Nb,Rd ∗ = 𝜒 ∗ Nc,Rd (10)
overall slenderness ratio is added by a ratio obtained from a distance ( √ )
between the batten panels and the minimum radius of gyration of the 2
𝜒 ∗ = 1∕ 𝛷 + 𝛷2 − 𝜆 (11)
full unreduced cross-sectional area of an individual shape in a built-up
16
Fig. 23. Comparison between the obtained results from the finite element and analytical methods (the EN1993-1-1, clause 6.3.1 [7] incorporated with the recommended slenderness
ratio from the AISI S100-16 [6]) for the battened columns with boundary conditions: (a) fixed and (b) pinned.
Fig. 24. Comparing axial buckling resistance obtained from the finite element models and the EN1993-1-1, clause 6.3.1 [7] incorporated with the recommended slenderness ratio
from the AISI S100-16 [6] for the battened columns with boundary conditions: (a) fixed and (b) pinned.
[ ( ∗ ) ]
∗2
𝛷∗ = 0.5 1 + 𝛼 𝜆 − 0.2 + 𝜆 (12) (Fig. 23a) and pinned (Fig. 23b) CFS built-up battened columns. As
shown in Fig. 23a, the fixed CFS built-up battened columns’ results
√
were above the buckling curve a. A similar result was seen for the
∗ 𝐴𝑒𝑓 𝑓 𝑓𝑦 { }
𝜆 = ; 𝑁𝑐𝑟 = min 𝑁𝑐𝑟,𝐹 ∗ , 𝑁𝑐𝑟,𝑇 , 𝑁𝑐𝑟,𝐹 𝑇 (13) pinned CFS built-up battened columns; however, the results for a few
𝑁𝑐𝑟 ∗
cases were slightly lower than the buckling curve a. The ratio obtained
𝜋 2 𝐸𝐴 from the finite element models over those buckling loads calculated
𝑁𝑐𝑟,𝐹 ∗ = ( )2 ( 𝑑 )2 (14)
𝐿𝑒 according to the EN1993-1-1, clause 6.3.1 [7] incorporated with the
𝑟
+ 𝑟𝑏𝑝
0 𝑖 recommended slenderness ratio from the AISI S100-16 [6] is shown
A comparison was made between the finite element and analytical in Fig. 24. As shown in Fig. 24a, most of the design buckling load
results (the EN1993-1-1, clause 6.3.1 [7] incorporated with the rec- considering buckling curve a is well predicted the axial load-bearing
ommended slenderness ratio from the AISI S100-16 [6]) for the fixed capacity with a mean ratio of 1.16. At the same time, a conservative
17
prediction of up to 23% is obtained considering the buckling curve load prediction by EN1993-1-1, clause 6.4.3 [7], as shown in Fig. 26.
b. Similar results were obtained for the pinned CFS built-up battened It is seen that the design buckling load predicted using the EN1993-1-1,
columns (Fig. 24b).
clause 6.4.3 [7], considering buckling curve b is safe.
According to Section 4.2, the reliability analysis was performed and
presented in Table 4 to evaluate the reliability of the methodology Table 5 lists detailed data for the reliability analysis for the analyti-
presented in the EN1993-1-1, clause 6.3.1 [7] incorporated with the
cal prediction according to EN1993-1-1, clause 6.3.1 [7]. The reliability
recommended slenderness ratio from the AISI S100-16 [6] for the CFS
built-up battened columns. It is seen that the reliability index is higher index value for the analytical approach, according to the EN1993-1-
than 2.5 for all situations. However, it is worth mentioning that the 1, clause 6.3.1 [7], shows values less than 2.5. These low values for
reliability index for the methodology based on the EN1993-1-1, clause the reliability index replicate the unreliable prediction provided by
6.3.1 [7] incorporated with the recommended slenderness ratio from
EN1993-1-1, clause 6.3.1 [7].
the AISI S100-16 [6] and considering buckling curve b is higher than
those considering buckling curve a, resulting in more reliable prediction √
𝐴𝑒𝑓 𝑓 𝑓𝑦
by considering buckling curve b. 𝜆∗∗ = (15)
𝑁𝑐𝑟,𝐵
4.5. Methodology in the EN1993-1-1, clause 6.4.3 [7] 1
𝑁𝑐𝑟,𝐵 = (16)
1 1
𝑁𝑐𝑟,𝑒𝑓 𝑓
+ 𝑆𝑣
The EN1993-1-1 [7] presents a design methodology for calculating
the design buckling resistance of the CFS built-up battened columns 𝜋 2 𝐸𝐼𝑒𝑓 𝑓
following clause 6.4.3. Note that the method available in EN1993-1- 𝑁𝑐𝑟,𝑒𝑓 𝑓 = (17)
𝐿2𝑒
1, clause 6.4.3 [7], is according to the recommendations provided by
the ECCS [65]. In general, this methodology, in particular, determines 𝐼𝑒𝑓 𝑓 = 0.5ℎ20 𝐴𝑐ℎ + 2𝜇𝐼𝑐ℎ (18)
the slenderness of the battened column, as presented in Eq. (15). The
critical load (𝑁𝑐𝑟,𝐵 ) is calculated according to Eq. (16), where the Elas- ⎧ 0, 𝜆 ≥ 150
tic flexural buckling force of the effective battened column is obtained ⎪
⎪ 𝜆
from Eq. (17) and assessed considering a reduced moment of inertia 𝜇 = ⎨2 − , 75 < 𝜆 < 150 (19)
75
(𝐼𝑒𝑓 𝑓 ). The reduced moment of inertia can be determined according to ⎪
⎪ 1.0, 𝜆 ≤ 75
Eqs. (18)–(20). Moreover, the shear stiffness of the built-up member ⎩
from the batten panel is determined using Eq. (21). √
Fig. 25 compares the obtained results from the finite element and 𝐿 𝐼1
𝜆 = ; 𝑖0 = ; 𝐼 = 0.5ℎ20 𝐴𝑐ℎ + 2𝐼𝑐ℎ (20)
analytical methods (EN1993-1-1, clause 6.4.3 [7]). For both fixed and 𝑖0 2 𝐴𝑐ℎ 1
pinned boundary conditions, the results followed the buckling curve a
24𝐸𝐼𝑐ℎ 2𝜋 2 𝐸𝐼𝑐ℎ
or between buckling curves a and b. However, the EN1993-1-1, clause 𝑆𝑣 = [ ] ≤ (21)
6.4.3 [7] recommended buckling curve b for calculating the reduction 2𝐼𝑐ℎ ℎ0 𝑏𝑏𝑝 2
𝑑𝑏𝑝 2 1 + 𝑛𝑏 𝐼𝑏𝑝 𝑑𝑏𝑝
factor. A comparison was made between the maximum axial load-
bearing capacity from the finite element model and design buckling
Fig. 25. Comparison between the obtained results from the finite element and analytical methods (EN1993-1-1, clause 6.4.3 [7]) for the battened columns with boundary conditions:
(a) fixed, (b) pinned.
18
Fig. 26. Comparing axial buckling resistance obtained from the finite element models and the EN1993-1-1, clause 6.4.3 [7] for the battened columns with boundary conditions:
5. Conclusions the axial capacity of the CFS built-up battened columns when
buckling curve a is considered. However, the reliability analysis
The compression behavior of various CFS built-up battened columns achieved a higher reliability index (𝛽 = 2.81 for fixed and 𝛽 =
(700 models) was investigated in this study. The modeling techniques 2.76 for pinned) when buckling curve b was used to predict the
were verified by comparing axial force vs. shortening displacement ob- axial capacity.
tained from numerical models and experimental specimens [10,30,61]. 4. The EN1993-1-1 [7] proposed a methodology for battened
The initial imperfection was defined using global and local buckling columns in clause 6.4.3. Buckling curve b also was suggested
modes. The optimal number of fastener rows per batten panel and the in the EN1993-1-1, clause 6.4.3 [7]. Its applicability for bat-
optimal distance between batten panels along the CFS built-up battened tened columns fabricated using 𝛴-shaped cold-formed steel pro-
columns were investigated through 100 finite element models. These files was investigated. The results showed that the predictions
suggestions were determined to achieve full or close to full composite provided by formulation in the EN1993-1-1, clause 6.4.3 [7],
action between two individual CFS profiles. Five different h/b ratios were conservative in general. However, the reliability index for
were considered in all steps of the investigation. After defining the its prediction (<2.5) represented clause 6.4.3 as an unreliable
optimal layout for the CFS built-up battened columns, a parametric procedure.
study was conducted using 600 finite element models to investigate the
accuracy of the available methodologies in the European Standards. 6. Future work
The findings are summarized as follows:
Although research studies were conducted on CFS built-up battened
1. The results recommended 30𝑖𝑚𝑖𝑛 as the distance between batten columns, many details are still missing to present a fully reliable de-
panels along the column, fastened using four rows of fasteners sign methodology. The compression behavior of CFS built-up battened
per batten panel. The results showed these values as optimal columns can be explored more when the batten panels are positioned in
to achieve the full or close to full composite action between the minor axis of the columns. Therefore, more experimental, numer-
two 𝛴-shaped cold-formed steel profiles. This recommended ical, and analytical are suggested for future studies to address more
distance between batten panels along the column agrees with remained details of the CFS built-up battened columns. It was con-
the previous finding [8,9]. cluded in Section 3.1 that increasing the axial load-bearing capacity due
2. Comparing finite element results and analytical procedures to the number of fasteners also included the effect of the batten panel’s
based on the EN1993-1-1, clause 6.3.1 [7] showed unconserva- stiffness. The contribution of the fasteners and the plate stiffness can
tive results for the CFS built-up section with batten panels when be addressed in more detail in future studies. Although four fastener
buckling curve a is considered. At the same time, the analytical rows per batten panel are suggested in this study, an investigation of
prediction considering buckling curve b was safe. Note that the the design procedure of batten panels is recommended in which the
analytical prediction taking the buckling curve-a into account required fasteners for the batten panel connection are determined by
represents a correct procedure, and only slightly unconservative computing the required resistance of connections between the profiles,
results were found for a few cases. the batten panels, and the shear resistance of each fastener.
3. The applicability of the EN1993-1-1, clause 6.3.1 [7], by adopt-
ing the slenderness from the AISI S100 [6] was proposed in [8]. CRediT authorship contribution statement
Its applicability for the presented battened columns fabricated
using 𝛴-shaped cold-formed steel profiles was investigated, re- Rohola Rahnavard: Visualization, Validation, Software, Methodol-
sulting in good accuracy. This study recommends the slenderness ogy, Investigation, Formal analysis, Data curation, Conceptualization,
from the AISI S100 [6] used in the EN1993-1-1, clause 6.3.1 [7], Writing – original draft, Writing – review & editing. Mahtabsadat
to predict the axial capacity of battened built-up columns. This Razavi: Writing – review & editing, Software, Formal analysis. Nader
paper showed that slenderness from the AISI S100 [6] used Fanaie: Writing – review & editing. Hélder D. Craveiro: Supervision,
in the EN1993-1-1, clause 6.3.1 [7] could accurately predict Writing – review & editing.
19
Declaration of competing interest [11] A.R. Dar, S. Vijayanand, M. Anbarasu, M.A. Dar, Cold-formed steel battened
built-up columns: Experimental behaviour and verification of different design
rules developed, Adv. Struct. Eng. 25 (2) (2022) 321–335, http://dx.doi.org/10.
The authors declare that they have no known competing finan-
1177/13694332211048006.
cial interests or personal relationships that could have appeared to [12] M. Adil Dar, Dipti Ranjan Sahoo, Sunil Pulikkal, Arvind K. Jain, Behaviour
influence the work reported in this paper. of laced built-up cold-formed steel columns: Experimental investigation and
numerical validation, Thin-Walled Struct. 132 (2018) 398–409, http://dx.doi.
Data availability org/10.1016/j.tws.2018.09.012.
[13] M. Adil Dar, Abhishek Verma, M. Anbarasu, Sze Dai Pang, A.R. Dar, Design of
cold-formed steel battened built-up columns, J. Construct. Steel Res. 193 (2022)
The authors do not have permission to share data. 107291, http://dx.doi.org/10.1016/j.jcsr.2022.107291.
[14] A.R. Dar, S. Vijayanand, M. Anbarasu, M.A. Dar, Cold-formed steel battened
Acknowledgments built-up columns: Experimental behaviour and verification of different design
rules developed, Adv. Struct. Eng. 25 (2) (2022) 321–335, http://dx.doi.org/10.
1177/13694332211048006.
This work is financed by national funds through Portuguese Foun- [15] M. Adil Dar, Dipti Ranjan Sahoo, Arvind K. Jain, Saurabh Sharma, Monotonic
dation for Science and Technology (FCT), under grant agreement tests and numerical validation of cold-formed steel battened built-up columns,
2021.06528.BD attributed to the 1st author and under the grant agree- Thin-Walled Struct. 159 (2021) 107275, http://dx.doi.org/10.1016/j.tws.2020.
ment 2020.03588.CEECIND attributed to the 4th author. 107275.
The authors gratefully acknowledge the Portuguese Foundation for [16] M. Adil Dar, N. Subramanian, M. Anbarasu, Ahmad Fayeq Ghowsi, P. Azmat Arif,
A.R. Dar, Testing and FE simulation of lightweight CFS composite built-up
Science and Technology (FCT) for its support under the framework
columns: Axial strength and deformation behaviour, Thin-Walled Struct. 167
of the research project PTDC/ECI- EGC/31858/2017 – INNOCFSCONC (2021) 108222, http://dx.doi.org/10.1016/j.tws.2021.108222.
– Innovative hybrid structural solutions using cold- formed steel and [17] Wilson Reyes, Andrés Guzmán, Evaluation of the slenderness ratio in built-
lightweight concrete’’, financed by FEDER funds through the Com- up cold-formed box sections, J. Construct. Steel Res. 67 (6) (2011) 929–935,
petitivity Factors Operational Programme-COMPETE and by national http://dx.doi.org/10.1016/j.jcsr.2011.02.003.
[18] Mohamed Dabaon, Ehab Ellobody, Khaled Ramzy, Nonlinear behaviour of built-
funds through FCT and D_INNOCFSCONC – ‘‘Demonstração de Solução
up cold-formed steel section battened columns, J. Construct. Steel Res. 110
Estrutural Híbrida Inovadora com Recurso a Aço-Enformado a Frio (2015) 16–28, http://dx.doi.org/10.1016/j.jcsr.2015.03.007.
e Betão Leve’’, funded by the call for starting projects and proof [19] Mohamed Dabaon, Ehab Ellobody, Khaled Ramzy, Experimental investigation
of concept projects UI-Transfer, with operation Code POCI-01-0246- of built-up cold-formed steel section battened columns, Thin-Walled Struct. 92
FEDER-181315 financed by COMPETE 2020. (2015) 137–145, http://dx.doi.org/10.1016/j.tws.2015.03.001.
[20] Khaled Ramzy, Mohamed Dabaon, Mahmoud El-Boghdadi, Experimental tests
This work was partly financed by FCT/MCTES through national
on built-up cold-formed steel section laced compression members, Thin-Walled
funds (PIDDAC) under the R&D Unit Institute for Sustainability and In- Struct. 172 (2022) 108882, http://dx.doi.org/10.1016/j.tws.2021.108882.
novation in Structural Engineering (ISISE), under reference [21] AS/NZS. Australian/New Zealand Standard, Cold-Formed Steel Structures,
UIDB/04029/2020. Standards Australia, Sydney, Australia, 1996.
[22] M.A. El Aghoury, A.H. Salem, M.T. Hanna, E.A. Amoush, Experimental inves-
tigation for the behaviour of battened beam–columns composed of four equal
slender angles, Thin-Walled Struct. 48 (9) (2010) 669–683, http://dx.doi.org/
10.1016/j.tws.2010.03.007.
[23] M.A. El Aghoury, A.H. Salem, M.T. Hanna, E.A. Amoush, Ultimate capacity of
battened columns composed of four equal slender angles, Thin-Walled Struct. 63
References (2013) 175–185, http://dx.doi.org/10.1016/j.tws.2012.07.019.
[24] M.A. El Aghoury, A.H. Salem, M.T. Hanna, E.A. Amoush, Strength of cold formed
[1] Ioannis Papargyriou, Iman Hajirasouliha, Jurgen Becque, Kypros Pilakoutas, battened columns subjected to eccentric axial compressive force, J. Construct.
Performance-based assessment of CFS strap-braced stud walls under seismic Steel Res. 113 (2015) 58–70, http://dx.doi.org/10.1016/j.jcsr.2015.04.008.
loading, J. Construct. Steel Res. 183 (2021) 106731, http://dx.doi.org/10.1016/ [25] M. Anbarasu, M. Adil Dar, Improved design procedure for battened cold-formed
j.jcsr.2021.106731. steel built-up columns composed of lipped angles, J. Construct. Steel Res. 164
[2] Ornella Iuorio, Vincenzo Macillo, Maria Teresa Terracciano, Tatiana Pali, Luigi (2020) 105781, http://dx.doi.org/10.1016/j.jcsr.2019.105781.
Fiorino, Raffaele Landolfo, Seismic response of Cfs strap-braced stud walls: [26] David C. Fratamico, Shahabeddin Torabian, Xi Zhao, Kim J.R. Rasmussen,
Experimental investigation, Thin-Walled Struct. 85 (2014) 466–480, http://dx. Benjamin W. Schafer, Experiments on the global buckling and collapse of built-
doi.org/10.1016/j.tws.2014.09.008. up cold-formed steel columns, J. Construct. Steel Res. 144 (2018) 65–80, http:
[3] Shuna Ni, Xia Yan, Matthew S. Hoehler, Thomas Gernay, Numerical modeling //dx.doi.org/10.1016/j.jcsr.2018.01.007.
of the post-fire performance of strap-braced cold-formed steel shear walls, [27] David C. Fratamico, Shahabeddin Torabian, Xi Zhao, Kim J.R. Rasmussen,
Thin-Walled Struct. 171 (2022) 108733, http://dx.doi.org/10.1016/j.tws.2021. Benjamin W. Schafer, Experimental study on the composite action in sheathed
108733. and bare built-up cold-formed steel columns, Thin-Walled Struct. 127 (2018)
[4] Luís Simões da Silva, Luís Carlos Silva, Trayana Tankova, Hélder David Craveiro, 290–305, http://dx.doi.org/10.1016/j.tws.2018.02.002.
Rui Simões, Ricardo Costa, Mario D’Aniello, Raffaele Landolfo, Performance [28] Jia-Hui Zhang, Ben Young, Compression tests of cold-formed steel I-shaped open
of modular hybrid cold-formed/tubular structural system, Structures 30 (2021) sections with edge and web stiffeners, Thin-Walled Struct. 52 (2012) 1–11,
1006–1019, http://dx.doi.org/10.1016/j.istruc.2021.01.066. http://dx.doi.org/10.1016/j.tws.2011.11.006.
[5] CEN, Eurocode 3, Design of Steel Structures-Part 1-3: General Rules- Supple- [29] Jia-Hui Zhang, Ben Young, Numerical investigation and design of cold-formed
mentary Rules for Cold-Formed Members and Sheeting, European Committee for steel built-up open section columns with longitudinal stiffeners, Thin-Walled
Standardization, Brussels, 2006. Struct. 89 (2015) 178–191, http://dx.doi.org/10.1016/j.tws.2014.12.011.
[6] AISI, North American Specification for the Design of Cold-Formed Steel Structural [30] Jia-Hui Zhang, Ben Young, Experimental investigation of cold-formed steel built-
Members, 2016. up closed section columns with web stiffeners, J. Construct. Steel Res. 147 (2018)
[7] CEN, Eurocode 3, Design of Steel Structures- Part 1-1: General Rules and Rules 380–392, http://dx.doi.org/10.1016/j.jcsr.2018.04.008.
for Building, European Committee for Standardization, Brussels, 2005. [31] Ben Young, Research on cold-formed steel columns, Thin-Walled Struct. 46 (7–9)
[8] Rohola Rahnavard, Hélder D. Craveiro, Rui A. Simões, Design of Cold-Formed (2008) 731–740, http://dx.doi.org/10.1016/j.tws.2008.01.025.
Steel Battened Built-Up Columns, Cold-Formed Steel Research Consortium [32] Mithum Peiris, Mahen Mahendran, Behaviour of cold-formed steel lipped channel
(CFSRC) Colloquium, 2022. sections subject to eccentric axial compression, J. Construct. Steel Res. 184
[9] Rohola Rahnavard, Hélder D. Craveiro, Luís Laím, Rui A. Simões, Rebecca (2021) 106808, http://dx.doi.org/10.1016/j.jcsr.2021.106808.
Napolitano, Numerical investigation on the composite action of cold-formed [33] Tekcham Gishan Singh, Konjengbam Darunkumar Singh, Design of perforated
steel built-up battened columns, Thin-Walled Struct. 162 (2021) 107553, http: cold-formed steel hollow stub columns using direct strength method, Thin-Walled
//dx.doi.org/10.1016/j.tws.2021.107553. Struct. 168 (2021) 108265, http://dx.doi.org/10.1016/j.tws.2021.108265.
[10] M. Adil Dar, Dipti Ranjan Sahoo, Arvind K. Jain, Abhishek Verma, Tests on CFS [34] Naoual Djafour, Soumia Kherbouche, Abdellatif Megnounif, Parametric study on
laced columns composed of plain channels: Behavior and design, J. Struct. Eng. cold-formed steel built-up columns using direct strength method, Structures 39
148 (5) (2022) http://dx.doi.org/10.1061/(ASCE)ST.1943-541X.0003329. (2022) 337–350, http://dx.doi.org/10.1016/j.istruc.2022.03.028.
20
[35] Qiu-Yun Li, Ben Young, Design of cold-formed steel built-up open section [50] Xuhong Zhou, Ming Chen, Experimental investigation and finite element anal-
members under combined compression and bending, Thin-Walled Struct. 172 ysis of web-stiffened cold-formed lipped channel columns with batten sheets,
(2022) 108890, http://dx.doi.org/10.1016/j.tws.2022.108890. Thin-Walled Struct. 125 (2018) 38–50, http://dx.doi.org/10.1016/j.tws.2018.01.
[36] Ziqi He, Haoyuan Cao, Xuhong Zhou, Qixiu Li, Benjamin W. Schafer, Perfor- 006.
mance of cold-formed C-sections reinforced by sleeves under concentric and [51] Francisco J. Meza, Jurgen Becque, Iman Hajirasouliha, Experimental study of
eccentric compression, J. Construct. Steel Res. 198 (2022) 107554, http://dx. cold-formed steel built-up columns, Thin-Walled Struct. 149 (2020) 106291,
doi.org/10.1016/j.jcsr.2022.107554. http://dx.doi.org/10.1016/j.tws.2019.106291.
[37] Kim J.R. Rasmussen, Mani Khezri, Benjamin W. Schafer, Hao Zhang, The [52] Francisco J. Meza, Jurgen Becque, Iman Hajirasouliha, Experimental study of
mechanics of built-up cold-formed steel members, Thin-Walled Struct. 154 (2020) the cross-sectional capacity of cold-formed steel built-up columns, Thin-Walled
106756, http://dx.doi.org/10.1016/j.tws.2020.106756. Struct. 155 (2020) 106958, http://dx.doi.org/10.1016/j.tws.2020.106958.
[38] M. Abbasi, M. Khezri, K.J.R. Rasmussen, B.W. Schafer, Elastic buckling analysis [53] Jun Ye, Iman Hajirasouliha, Jurgen Becque, Experimental investigation of local-
of cold-formed steel built-up sections with discrete fasteners using the compound flexural interactive buckling of cold-formed steel channel columns, Thin-Walled
strip method, Thin-Walled Struct. 124 (2018) 58–71, http://dx.doi.org/10.1016/ Struct. 125 (2018) 245–258, http://dx.doi.org/10.1016/j.tws.2018.01.020.
j.tws.2017.11.046. [54] Seyed Mohammad Mojtabaei, Jurgen Becque, Iman Hajirasouliha, Structural size
[39] Iveta Georgieva, Luc Schueremans, Lincy Pyl, Composed columns from cold- optimization of single and built-up cold-formed steel beam-column members, J.
formed steel Z-profiles: Experiments and code-based predictions of the overall Struct. Eng. 147 (2021) 04021030, http://dx.doi.org/10.1061/(ASCE)ST.1943-
compression capacity, Eng. Struct. 37 (2012) 125–134, http://dx.doi.org/10. 541X.0002987.
1016/j.engstruct.2011.12.017. [55] S.J. Qadir, V.B. Nguyen, I. Hajirasouliha, B. Cartwright, M.A. English, Optimal
[40] Iveta Georgieva, Luc Schueremans, Lincy Pyl, Lucie Vandewalle, Experimen- design of cold roll formed steel channel sections under bending considering
tal investigation of built-up double-z members in bending and compression, both geometry and cold work effects, Thin-Walled Struct. 157 (2020) 107020,
Thin-Walled Struct. 53 (2012) 48–57, http://dx.doi.org/10.1016/j.tws.2011.12. http://dx.doi.org/10.1016/j.tws.2020.107020.
017. [56] Jurgen Becque, Kim J.R. Rasmussen, Experimental investigation of local-overall
[41] Iveta Georgieva, Luc Schueremans, Lincy Pyl, Lucie Vandewalle, Numerical study interaction buckling of stainless steel lipped channel columns, J. Construct. Steel
of built-up double-z members in bending and compression, Thin-Walled Struct. Res. 65 (8–9) (2009) 1677–1684, http://dx.doi.org/10.1016/j.jcsr.2009.04.025.
60 (2012) 85–97, http://dx.doi.org/10.1016/j.tws.2012.07.005. [57] Abaqus User’s Manual, 2011.
[42] I. Georgieva, L. Schueremans, L. Vandewalle, L. Pyl, Design of built-up cold- [58] Rohola Rahnavard, Hélder D. Craveiro, Rui A. Simões, Luís Laím, Aldina
formed steel columns according to the direct strength method, Procedia Eng. 40 Santiago, Buckling resistance of concrete-filled cold-formed steel (CF-CFS) built-
(2012) 119–124, http://dx.doi.org/10.1016/j.proeng.2012.07.066. up short columns under compression, Thin-Walled Struct. 170 (2022) 108638,
[43] Paweł Lorkowski, Bronisław Gosowski, Experimental and numerical research of http://dx.doi.org/10.1016/j.tws.2021.108638.
the torsion problem of built-up steel columns laced in a single plane, Eng. Struct. [59] Hélder D. Craveiro, João Paulo C. Rodrigues, Aldina Santiago, Luís Laím, Review
160 (2018) 566–580, http://dx.doi.org/10.1016/j.engstruct.2018.01.049. of the high temperature mechanical and thermal properties of the steels used in
[44] Krishanu Roy, Chia Mohammadjani, James B.P. Lim, Experimental and numerical cold formed steel structures – The case of the S280 Gd+Z steel, Thin-Walled
investigation into the behaviour of face-to-face built-up cold-formed steel channel Struct. 98 (Part A) (2016) 154–168, http://dx.doi.org/10.1016/j.tws.2015.06.
sections under compression, Thin-Walled Struct. 134 (2019) 291–309, http: 002.
//dx.doi.org/10.1016/j.tws.2018.09.045. [60] Hélder David da Silva Craveiro, Fire Resistance of Cold-Formed Steel
[45] T.C.H. Ting, K. Roy, H.H. Lau, J.B. Lim, Effect of screw spacing on behavior of ax- Columns (PHD thesis), ISISE-Depatment of Civil Engineering, University of
ially loaded back-to-back cold-formed steel built-up channel sections, Adv. Struct. Coimbra, Portugal, 2016.
Eng. 21 (3) (2018) 474–487, http://dx.doi.org/10.1177/1369433217719986. [61] Hélder D. Craveiro, Rohola Rahnavard, Luís Laím, Rui A. Simões, Aldina
[46] M. Muthuraman, R. Anuradha, P.O. Awoyera, R. Gobinath, Numerical simulation Santiago, Buckling behavior of closed built-up cold-formed steel columns under
and specification provisions for buckling characteristics of a built-up steel column compression, Thin-Walled Struct. 179 (2022) 109493, http://dx.doi.org/10.
section subjected to axial loading, Eng. Struct. 207 (2020) 110256, http://dx. 1016/j.tws.2022.109493.
doi.org/10.1016/j.engstruct.2020.110256. [62] Minh Toan Huynh, Cao Hung Pham, Gregory J. Hancock, Experimental behaviour
[47] Jessica Whittle, Chris Ramseyer, Buckling capacities of axially loaded, cold- and modelling of screwed connections of high strength sheet steels in shear,
formed, built-up C-channels, Thin-Walled Struct. 47 (2) (2009) 190–201, http: Thin-Walled Struct. 146 (2020) 106357, http://dx.doi.org/10.1016/j.tws.2019.
//dx.doi.org/10.1016/j.tws.2008.05.014. 106357.
[48] Yuanqi Li, Yinglei Li, Shukun Wang, Zuyan Shen, Ultimate load-carrying capacity [63] Guan Quan, Jun Ye, Wenchao Li, Computational modelling of Cold-formed steel
of cold-formed thin-walled columns with built-up box and I section under axial lap joints with screw fasteners, Structures 33 (2021) 230–245, http://dx.doi.org/
compression, Thin-Walled Struct. 79 (2014) 202–217, http://dx.doi.org/10.1016/ 10.1016/j.istruc.2021.04.062.
j.tws.2014.02.003, m. [64] CEN, Eurocode 3, Design of Steel Structures-Part 1-5: Plated Structural Elements,
[49] Hélder D. Craveiro, João Paulo C. Rodrigues, Luís Laím, Buckling resistance European Committee for Standardisation, Brussels, Belgium, 2006.
of axially loaded cold-formed steel columns, Thin-Walled Struct. 106 (2016) [65] CECM ECCS, European Recommendations for Steel Construction, ECCS, Brussels,
358–375, http://dx.doi.org/10.1016/j.tws.2016.05.010. 1978.
21

Evaluation of The Composite Action of Cold Formed Steel Buil - 2023 - Thin Walle

Uploaded by

Copyright:

Available Formats

Evaluation of The Composite Action of Cold Formed Steel Buil - 2023 - Thin Walle

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Evaluation of The Composite Action of Cold Formed Steel Buil - 2023 - Thin Walle

Uploaded by

Copyright:

Available Formats

Thin-Walled Structures 183 (2023) 110390

Contents lists available at ScienceDirect

Evaluation of the composite action of cold-formed steel built-up battened

ARTICLE INFO ABSTRACT

1. Introduction knowledge of how these sections behave is still restricted. Because

Notation 𝛷𝑐 Safety factors for the available axial

2.3. Numerical modeling

Fig. 2. The battened columns with varied ℎ∕𝑏 ratio.

Fig. 3. Geometry details of individual CFS profiles.

Fig. 4. Geometric details of the battened column.

was defined according to the stress–strain curve presented for

2.3.3. Imperfection, boundary conditions, and loading

Fig. 6. Finite element modeling.

Comparing Figs. 16 and 17 shows that the differences between the

Number of models # 300 300 300 300

Number of models # 300 300 300 300

Number of models # 300 300 300 300

You might also like